Judgments of personality are inferences about traits that underlie consistent, coherent patterns of behavior across diverse situations. Such judgments are relevant to mental health from both the clinician's and client's perspective. From the clinician's perspective, a better understanding of the bases of accurate interpersonal judgment is relevant to improvements in clinical assessment, evaluation, diagnosis, and subsequent treatment. From the client's perspective, many problems in living and several of the major personality disorders are directly relevant to or even stem directly from deficiencies in the accuracy of social perception. Patterns of paranoia, hostility, fearfulness, shyness, narcissism and others are produced, exacerbated and sometimes fundamentally characterized by inaccurate social perception. Research on accurate personality judgment over the past decade has settled several important issues, and opened others. Some of the unsettled issues are addressed by the PI's Realistic Accuracy Model (RAM, Funder, 1995), which describes how accurate personality judgment is the result of a social psychological process by which relevant information becomes available to and then is detected and correctly utilized by a judge. RAM yields hypotheses concerning the effect of informational context and individual differences among judges on the accuracy of personality judgment. To test these hypotheses and address other issues, 144 undergraduate subjects of both sexes and varied ethnicity will interact in one of 4 1-hour experimental contexts, and another 4-hour context. These contexts are designed to vary in the kind and amount of personality-relevant information that becomes available in them. Afterwards, subjects will provide their judgments of each others' personalities. The accuracy of these judgments will be assessed against and correlated with a wide range of measures of personality and mental health derived from self-reports, informants' reports, assessments by mental health professionals, and direct behavioral observations. Hypotheses predict differences in accuracy across experimental conditions as well as mediating individual differences in information pickup and judgmental ability within conditions. Data will be analyzed using profile analyses, item analyses, the social relations model, and structural equation modeling.